### Table 1 Characteristics of explicit, emergent, and embedded foresight exercises Explicit Emergent Embedded

"... In PAGE 10: ... Hence, emergent foresight may be susceptible to corporatism, which suggests that it should not be ascribed a dominant role in public policy. These views on different foresight modes are summarized in Table1 . Further comparisons are provided in Table 2.... ..."

### Table 11 Characteristics of explicit, emergent and embedded foresight processes, adapted from Salmenkaita and Salo (2003) Explicit Emergent Embedded

2006

### Table 1. Characteristics of explicit, emergent, and embedded foresight exercises

2004

"... In PAGE 13: ... Hence, emergent foresight may be susceptible to corporatism, which suggests that it should not be ascribed a dominant role in public policy. These views on different foresight modes are summa- rized in Table1 . Further comparisons are provided in Table 2.... In PAGE 13: ... Further comparisons are provided in Table 2. [INSERT Table1 and Table 2 AROUND HERE] Preconditions for Emergent Foresight Emergent foresight activities are unlikely to occur in the absence of several contextual prerequisites. To begin with, these activities build on existing networks of stakeholders who are interdependent in the sense that the future opportunities of each are influ- enced by the actions of others; thus, in economic terms, oligopolistic industries are more prone to engage in emergent foresight.... ..."

Cited by 4

### Table VIII. Time and memory requirements for different edge-caching schemes. All times are in seconds. Explicit caching refers to the algorithm of Reitsma and Pollard 2004. The + notation is used as we replicated only the first parts of their algorithm for comparative purposes; some additional computation beyond the amount timed is required by their algorithm. The caching scheme used allows a tradeoff between space and time requirements for the evaluation, although a simple caching scheme achieved most of the benefits of full caching while keeping the memory footprint low: the running time of our on-demand algorithm is within a small constant factor of what could be achieved if the embedded graph were built explicitly, while the memory required grows at an asymptotically lower rate. (The Metrics column corresponds to the relative performance one would expect when using the embedded graph (e.g., searching for paths).)

2001

Cited by 1

### Table VIII. Time and memory requirements for different edge-caching schemes. All times are in seconds. Explicit caching refers to the algorithm of Reitsma and Pollard 2004. The + notation is used as we replicated only the first parts of their algorithm for comparative purposes; some additional computation beyond the amount timed is required by their algorithm. The caching scheme used allows a tradeoff between space and time requirements for the evaluation, although a simple caching scheme achieved most of the benefits of full caching while keeping the memory footprint low: the running time of our on-demand algorithm is within a small constant factor of what could be achieved if the embedded graph were built explicitly, while the memory required grows at an asymptotically lower rate. (The Metrics column corresponds to the relative performance one would expect when using the embedded graph (e.g., searching for paths).)

2001

Cited by 1

### Table 3: Allowed and Disallowed Constraining Binary Impositions [JONE96] The constraining binary relationship in our example is allowable according to the EBP rule. Figure 7 shows the ternary relationship from Figure 6 with the explicit binary relationship imposed, the resulting maximum cardinality constraints of the embedded binary relationships, and a possible instance table to reflect the instance triples of the diagram. Important to our analysis of structural validity is the changes that occur to the embedded binary relationships due to the imposition. The embedded relationship between Project and Budget follows the constraining M:1 relationship as expected but additional the embedded relationship between Project and Team also changes to M:1 because of the imposition. The embedded relationships change because of the additional binary FD(s) imposed on the ternary relationship and the additional binary FDs that may be derived from the imposition on the ternary relationships.

"... In PAGE 12: ... Their rule states: For any given ternary relationship, a binary relationship cannot be imposed where the binary cardinality is less than the cardinality specified by the ternary, for any specific entity. Table3 summarizes the allowable and disallowable binary impositions on the different cardinality... In PAGE 15: ... The first step again is to analyze the ternary relationship with the constraining relationships imposed on it. R(xy) is a M:1 constraining relationship and from Table3 or the EBP rule we conclude it is a valid imposition on a M:1:1 ternary relationship. Therefore the ternary is structurally valid.... ..."

Cited by 2

### Table 10. Explicit de nition constants/operators of ACPsatIp in ACPnsatIp

2001

"... In PAGE 16: ... Besides, the operator abs has an element of R instead of an element of F as its rst argument and the operator R has a subset of R instead of a subset of F as its rst argument. Explicit de nitions of these constants and operators in ACPsatIp are given in Table10 . Notice that the Table 10.... In PAGE 17: ... Theorem 1. The explicit de nitions given in Table10 induce an embedding of ACPsatIp in ACPnsatIp. Proof.... In PAGE 17: ... Proof. In order to prove that the explicit de nitions given in Table10 induce an embedding of ACPsatIp in ACPnsatIp, we have to check that the axioms of ACPsatIp are derivable for closed terms from the axioms of ACPnsatIp and the explicit de nitions. Using Lemmas 3, 4 and 5 given below, it is straightforward to check the derivability of the axioms of ACPsatIp.... ..."

Cited by 4

### Table 6. Timings of group operations with ARMulator ARM7TDMI@80MHz (explicit formulae)

2003

"... In PAGE 11: ...ase of HECC. For genus-3 curves the polynomial h(x) equals one. The group orders range from 2162 to 2190. Table6 presents timings for divisor addition, divisor doubling and scalar multiplication on the ARMulator. To our knowledge these are the first published timings for HECC on an embedded processor.... ..."

Cited by 35

### Table 6. Timings of group operations with ARMulator ARM7TDMI@80MHz (explicit formulae)

"... In PAGE 11: ...ase of HECC. For genus-3 curves the polynomial h(x) equals one. The group orders range from 2162 to 2190. Table6 presents timings for divisor addition, divisor doubling and scalar multiplication on the ARMulator. To our knowledge these are the first published timings for HECC on an embedded processor.... ..."